| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
|
|
|
|
| |
* Add a portable interface to strerror
* Add Status::FromErrno
* Add strerror_test.cc to the Xcode project
* Use glibc feature selection macros instead of return-type overloads
* Fix tensorflow references
|
| |
|
|
|
|
|
|
| |
* Remove extraneous firebase_firestore_util_async_queue target
* Remove unimplemented declaration in string_util.h
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Don't rely on specialization failure to determine when std::hash is unavailable.
Instead manually declare the conditions under which std::hash should be
defined.
* Fix detection of Objective-C classes in Xcode < 9.2
std::is_base_of<NSObject, NSString>{} is false there so the overloads
defined for Objective-C types weren't getting enabled.
* Add explicit tests for StringFormat using Objective-C objects
* Add explicit tests for HasStdHash
|
| |
|
| |
|
|
|
|
|
|
|
| |
* Add direct support for formatting Objective-C objects
* Rewrite log.h
* Convert FSTWarn to LOG_WARN
* Convert FSTLog to LOG_DEBUG
* Remove FSTLogger
|
|
|
|
|
|
| |
* Add HARD_ASSERT
* Use HARD_ASSERT
* Remove FIREBASE_ASSERT
* Remove StringPrintf
|
|
|
|
| |
* Add StringFormat
* Use StringFormat
|
|
|
|
|
|
| |
* fix lint to run copybara
* address changes
|
|
|
|
|
| |
FSTDispatchQueue now doesn't contain any logic of its own and instead
just passes through all method calls to AsyncQueue (backed by an
ExecutorLibdispatch).
|
|
|
|
|
| |
This should prevent weird bugs due to system time readjustments. Because
operations are scheduled relative to now, the fact that the resulting
timepoint isn't in Unix epoch shouldn't matter.
|
|
|
|
|
|
|
|
| |
dispatch_queue_t is defined differently in libdispatch depending on
whether the library header is being include from Objective-C (or
Objective-C++) code, or else from C or C++ code. Make sure that all
source files in Firestore that include executor_libdispatch.h are
compiled in the same mode (Objective-C++) to avoid linker errors.
|
|
|
|
|
|
|
|
|
|
|
|
| |
AsyncQueue is a queue that executes given operations asynchronously,
enforcing that only a single operation is executing at any given time,
and that in-progress operations don't spawn more operations. The actual
execution is delegated to a platform-specific executor. Executor is an
interface for a FIFO queue that executes given operations serially. Two
implementations of Executor, one using libdispatch and the other using
C++11 standard library, are provided.
AsyncQueue is not used anywhere in the code base at this point.
|
|
|
|
|
| |
This is good enough to make it possible for the new C++ code to
interoperate with existing Objective-C code where `-hash` is required if
you override `-isEqual:`.
|
|
|
|
|
|
| |
* Add a simple range adapter.
* Add SortedMap::keys
* Add SortedMap::keys_from
* Add SortedMap::keys_in
|
|
|
|
|
|
| |
* Standardize method ordering across sorted maps
* Add SortedMap::find
* Add SortedMap::find_index
* Add SortedMap::contains
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add a .clang-tidy configuration for Firestore C++
* Fix clang-tidy warnings
* typedef -> using
* const ref + rvalue ref -> pass by value
* NULL -> nullptr
* remove useless default initializations
* remove useless const value-type parameter declarations (definitions
can still use them)
* use auto instead of repeating types in a cast
* Fix typos
* Address use of static method through instance warnings
* Address use after move warnings
|
| |
|
|
|
| |
Remove a duplicate macro definition and clean up an extra const clang-tidy complains about.
|
|
|
|
| |
Since we now include cmath rather than math.h, isnan is now named
std::isnan.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* lint.sh now lints Objective-C++ too
* cpplint checks system-style includes that should be user-style
This prevents it from recognizing project sources as if they were C
system headers and then complaining that they're in the wrong place.
* cpplint checks #imports and #includes
* cpplint checks that C++ system headers aren't #imported
* cpplint checks for C system headers that could be C++ system headers
* cpplint checks that Objective-C sources include their headers
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
* Prepare for TreeSortedMap
* Factor out SortedMapBase
* Move ArraySortedMap to impl
* Factor out SortedMap testing utilities
* Add a minimal TreeSortedMap
* Add the public SortedMap type
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Change CredentialsProvider::TokenListener to use StatusOr
Rather than a token plus error code/msg.
* Eliminate the concept of an invalid Token
Instead, we'll just use StatusOr<Token>. Note that unauthenticated
tokens are handled as a special case; they're created via:
Token::Unauthenticated()
and are otherwise "valid", though attempting to retrieve the raw token
on one of these tokens will cause an assertion failure.
|
|
|
|
| |
* Pull in status files from tensorflow
* Add missing dependency to immutable library
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* naively remove FSTPath import and source/test files.
* port FieldPath, part I
* port FieldPath, part II
* port ResourcePath, part I
* port ResourcePath, part II
* the grand commit to fix build errors
* use testutil:: helper instead of those from FSTHelpers
* fix test and lint
* use c_str in errmsg directly
* fix
* fix
* make code clean
* fix integration test I missed
* fix to avoid naming collision in preprocessor
* address changes
* address changes
* address changes
* fix: fieldMask are actually shared with different context.
* address changes
* address changes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Fix ArraySortedMap.ChecksSize test (when NDEBUG is enabled)
assert() is a noop when NDEBUG is defined (i.e. during production
builds) so the size assertion check doesn't occur. Assuming we want this
check even during prod use, I've switched assert to FIREBASE_ASSERT.
* s/FIREBASE_ASSERT/FIREBASE_DEV_ASSERT/g in ordered_code.cc
This keeps the behaviour in line with the original (which purposely
differs if NDEBUG is defined or not.) Discovered due to the test suite
(correctly) breaking during a prod build.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* lazy replacing FST(Firebase)CredentialsProvider by (Firebase)CredentialsProvider
* lazy replacing FSTUser by User
* adding error-code parameter to TokenListener
* actually use const user& instead of pointer; also add an error util
* add HashUser and pass into the unordered_map
* use User in test
* use c++ CredentialsProvider and subclass in test
* fix unit test
* use explicit capture in lambda instead of capture all by reference
* cache currentUser explicitly when reset sync engineer test driver
* objc object should be captured by value in lambda
* replacing Auth/FSTUser by C++ auth implementation
* address changes
* replacing FSTGetTokenResult by C++ Token implementation
* address changes
* fix unintentional change in merging
* patch the change in objc Auth up-stream
* somehow, the lambda-version of set-user-change-listener does not work... fallback to block
* address changes
* fix another const& v.s. dispatch bug
* fix more const& v.s. dispatch bug zxu123 committed
* fix a bad sync line
* address changes
* address change
* address change
* fix upstream change from merge
* fix upstream changes
* Suggested fixes for cpp/port_auth (#846)
* Get rid of MockDatastore factory
This avoids the need to statically allocate (and leak) a credentials
provider
* Use absl::make_unique
std::make_unique technically does not exist until C++14.
* #include <utility> for std::move
* Use std::future for the initial user
* fix style
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* replacing Auth/FSTUser by C++ auth implementation
* address changes
* replacing FSTGetTokenResult by C++ Token implementation
* address changes
* address changes
* fix another const& v.s. dispatch bug
* fix more const& v.s. dispatch bug zxu123 committed
* fix
* passing by value in callback
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Implement firestore/auth/user
* add user to project and some fixes
* implement firestore/auth/{credentials_provider,empty_credentials_provider}
* implement firestore/auth/firebase_credentials_provider
* refactoring firebase_credentials_provider and add (disabled but working) unit test
* add auth test to project
* address changes
* small fix to style and project
* fix the firebase_credentials_provider_test
* fix style
* address changes
* revert the change to static mutex_
* remove my custom plist path
* fix style
* address changes
* refactoring FirebaseCredentialsProvider to fix the issue w.r.t. auth global dispatch queue
* add /*force_refresh=*/ tag to bool literal for style purpose
* Use a shared_ptr/weak_ptr handoff on FirebaseCredentialsProvider (#778)
* Revert "refactoring FirebaseCredentialsProvider to fix the issue w.r.t. auth global dispatch queue"
This reverts commit 87175a4146267d403a774f138b85f8d3b532fa4b.
* Use a shared_ptr/weak_ptr handoff on FirebaseCredentialsProvider
This avoids any problems with callsbacks retaining pointers to objects
destroyed by a C++ destructor
|
|
|
| |
Similar to Objective-C, FieldPath and ResourcePath share most of their interface (and implementation) by deriving from BasePath (using CRTP, so that factory methods in BasePath can return an instance of the derived class).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
iOS code (#738)
* implement Firestore DatabaseInfo in C++
* temporary stash changes; blocking on the massive renaming of .m to .mm
* add database_info_test to project
* finish port DatabaseId and fix style, modular fixing DatabaseInfo
* port DatabaseInfo
* remove FSTDatabase{ID,Info} and their tests from project
* fix unit test
* use namespace alias
* use namespace alias, leftover
* address more changes
* refactoring to use raw pointer instead of value for property
* address changes
* remove self->
* fix style
* remove the name suffix Alloc
* fix a bug
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Import iterator_adapters from google3
* Remove -Wconversion which is annoyingly hard to satisfy
* Strip dependency on absl_container from iterator_adapters_test
* Format and lint iterator_adaptors
* More flexible copyright checking in Travis
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Move -fvisibility-inlines-hidden from common_flags to cxx_flags
This option isn't supported by C (and causes the build to fail under
at least rodete).
Manifested error was:
```
-- Looking for include file openssl/rand.h
-- Looking for include file openssl/rand.h - not found
CMake Error at core/src/firebase/firestore/util/CMakeLists.txt:94 (message):
No implementation for SecureRandom available.
-- Configuring incomplete, errors occurred!
```
which is completely misleading. :(
* Fix exception include for std::logic_error
Was causing compiler error on (at least) rodete.
(http://en.cppreference.com/w/cpp/error/logic_error suggests that
stdexcept is the correct header.)
* Remove 'const' from vector contents.
vectors cannot contain const objects as the contained objects are
required to be assignable and copy constructable. (Not *quite* strictly
true; since c++11, this requirement now depends on the operations
performed on the container. But my compiler objects at any rate.)
http://en.cppreference.com/w/cpp/container/vector
https://stackoverflow.com/questions/8685257/why-cant-you-put-a-const-object-into-a-stl-container
* Add brackets as suggested (required) by my compiler.
* Add missing include
For LLONG_MIN, etc.
* Enable gnu extension to c++11 for firestore/util *test*.
We disable them by default (in cmake/CompilerSetup.cmake) but our tests
requires hexidecimal floating point, which is not supported in c++11
(though is supported in C++17, gnu++11, and others).
http://en.cppreference.com/w/cpp/language/floating_literal
https://bugzilla.redhat.com/show_bug.cgi?id=1321986
* Fix printf format for uint64_t
http://en.cppreference.com/w/cpp/types/integer
https://stackoverflow.com/questions/9225567/how-to-print-a-int64-t-type-in-c
* Allow 0 length printf template strings in tests.
* Get rid of a multi line comment.
The backslash is apparently interpreted by the compiler cause the next
line to be ignored too. In this case, the next line is (currently) a
comment, and would be ignored anyways. (But that doesn't stop the
compiler from yelling.)
* Run ./scripts/style.sh
|
|
|
|
| |
* normalize bits
* normalize ordered_code
|
|
|
|
| |
* refactoring string_util
* port string_util to iOS
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reimplements our comparison functions as C++ Comparators and then provides compatibility shims for interoperating with existing Objective-C usage.
A few specialized comparators aren't suitable for porting but only have a single usage (e.g. CompareBytes for comparing NSData * instances). In these cases I've moved them into the caller.
* Use int32_t for typeof(ID) in FSTDocumentReference
* Migrate callers of FSTComparison.h to Objective-C++
* Port comparison to C++
* Migrate usages of FSTComparison.h to C++ equivalents
* Remove FSTComparison
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add CMake platform detection logic for SecureRandom
Now only builds secure_random_arc4random.cc if available.
Remove firebase/firestore/base/port.h. Nothing else was in that
directory.
* Add a SecureRandom implementation that uses OpenSSL
This is usable on Linux, Windows, and Android
* Properly check return from RAND_bytes
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Rewrite cc_test to take named arguments
Cut down on build file verbosity by having cc_test take SOURCES and
DEPENDS. The separate invocation of target_link_libraries is no longer
necessary.
* Add a cc_library rule to parallel cc_test
This cuts down on build file verbosity.
* Automatically add OBJC_FLAGS to cc_libraries if applicable
* Exclude platform-specific libraries from 'all'
This is makes it possible to declare this kind of library
unconditionally. Usage within a test or as a dependency will actually
trigger building.
* Restore secure_random_test.cc; clean-up comments
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* implement FieldValue for null and boolean.
* refactoring to use union type instead of poly
* refactor using union design intead of poly
* refactoring to use anonymous union and fix styles
* small fix
* add field_value_test to the project
* fix warning of cmake and fix style
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Misc style.sh fixes
* Allow test-only to use a revision; to check your changes since
master:
./scripts/style.sh test-only master
* Avoid diffing deleted files
* 80 columns
* Fix C++ lint errors
|
|
|
|
|
|
| |
* Enable warnings when building with GCC or clang
* Fix warnings
|
|
|
|
| |
Run style.sh (w/ clang-format version 6)
|
|
|
|
|
|
|
|
|
| |
* implement C++ assert (stdio, apple)
* Update tests for firebase_firestore_util renames
* renaming `assert.h` to `firebase_assert.h`
* refactoring to a common `WrapNSStringNoCopy()`
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Port StringPrintf from //base.
Prefer this to approaches based on variadic templates. While the variadic
template mechanisms are strictly safer, they result in binary bloat we
can't afford.
This is essentially the same StringPrintf previously open sourced as a
part of protobuf, though updated for C++11 which saves a copy and a
temporary buffer on the heap.
* Add abseil as a subdirectory of Firestore
This saves having to redefine all the libraries that abseil defines as
imported libraries.
* Rename firebase_firesture_util_log_* targets
Cut the log out of the name to reflect that these will get more
components besides just logging.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Rename FIREBASE_BINARY_DIR to FIREBASE_INSTALL_DIR
Make this consistent with the outer superbuild and also make the
association with CMAKE_INSTALL_PREFIX more obvious.
* Build and test log_stdio separate from the rest of util
This is in preparation for adding a test for log_apple
* Build and test log_apple under CMake
Also add notes about how FIRLogger's debug mode can break this test
* Refactor log_apple to cut down duplicate switch statements
There's also a slight reduction in final binary size.
|